Airbase Damage Representation Improvements – Oct 2005
The airbase damage model for conventional munitions has been updated and now uses the algorithm described in this document. This damage model is limited to determining the number of aircraft destroyed at the base and the time the airbase will return to an operational state. Special notes are:
The aircraft are either destroyed or left intact; aircraft damage is not considered
No specific airbase facilities (towers, etc) are discretely tracked for damage in this model. Instead the overall base operations are tracked in terms of the ability to generate sorties. And sortie generation is binary; that is the base can either generate sorties or it can’t (unlike the bio/chem model, which actually reduces sortie generation rate). This is done so eliminate any requirement for the analyst to laydown and define airbase facilities for each airbase during scenario generation. While JFORCES incorporates detailed models to accommodate this (e.g. the IBDP), this capability is not part of the baseline because the workload is significant and has been deemed not to support the thrust of most analyses. But the airbase attrition model has been kept modular for replacement with a more detailed representation when required.
The are two other airbase damage models based on the effects of biochem and nuclear attacks. These models have not been changed. For reference, the airbase degradation resulting from biological/chemical attack is still modeled as a reduction of sortie rates from the base that is cranked into the ATO module. And the nuclear damage model is still binary; that is the base is either killed or left intact.
When an airbase is attacked with conventional munitions, the damage module can call a specific airbase damage model. Currently, the main controlling routine for conventional damage determination is bomb_em.F, found in the sim/environ directory. The pertinent changes to this algorithm are as follows:
When the algorithm identifies targets in the detonation zone it identifies any airbases. The algorithm operates differently based upon the mission targeting inputs:
If the mission was targeted against the airbase the airbase_attrition_ routine is called (described below) and all attrition computations for all bombs dropped in this mission are performed by this routine.
If this mission is not specifically targeted against any asset, the algorithm Monte Carlo’s that the attack focused exclusively on airbase elements at 80%. The reason that we didn’t automatically assume that the mission was always focused on the airbase is this approach permits the attack of ancillary elements including air defense assets around the base.
If this mission is specifically targeted against an asset other than the airbase, then the attack is focused on that asset.
If this mission was determined to focus exclusively on the airbase (either by mission specification or the Monte Carlo draw on unspecified missions), all further damage calculations are performed by the airbase_attrition_ algorithm, currently found in the sim/engment/survivability.c file (and described below). If the mission is not considered focused on the airbase (either by specific mission directives or according to the Monte Carlo draw), the standard approach of looping through all targets in the damage zone of the munitions drop(s) is performed to find all damaged assets. This calculation considers the spread of munitions based on targeting criteria, warhead guidance options, and CEP. The airbase is still considered as one of the candidate damage targets and could be damaged as a result of this calculation. Should one or more bombs be evaluated as attacking the base in this loop the airbase_attrition_ routine is again called. The change in this case is the airbase (and associated aircraft) damage estimate is based on the specific bomb(s) that are determined to hit the airbase, not the entire bombload delivered in this mission. The other bombs are mapped to damage against non-airbase targets.
As described above, the airbase_attrition_ routine is called only for damage calculations to airbases. The code is currently found in the sim/engment/survivability.c file. An overview of this algorithm is:
The system determines whether the munition dropped is tailored to destroy runways. This will adjust the airbase downtime, calculated later.
The algorithm then determines an adjustment to the probability of killing aircraft based on a estimate of the probability that the aircraft will be located in revetments or hangers. There are several hardcoded values here that the analyst should be aware of. Details are:
The total number of aircraft at the base is determined.
If the airbase has more revetments than aircraft, all aircraft except 5 are deemed to be in the revetments. The reason that 5 are exempt is this a WAG for aircraft that might be exposed at any time for airbase operations (e.g. taxiing). If there are fewer aircraft than revetments all revetments except 3 are deemed to have aircraft in them. Again, 3 is a WAG used to consider aircraft in transition for airbase operations.
The remaining aircraft (those not in revetments) are compared to hanger space. If there’s sufficient hanger space for all of these aircraft all except 3 are considered to be in hangers. Again, 3 is a WAG to represent aircraft in transition. If there’s more aircraft than hanger space the hangers are deemed to be filled to 2 less than capacity (thus if there’s room for 20 aircraft, no more than 18 spaces will be used). Again, this WAG represents operations and transition.
The total survivability improvement is based on the number of aircraft in revetments, hangers and exposed. The survivability factors for aircraft in hangers and revetments is as specified by the user for each airbase. These factors are used as a hardness multiplier to aircraft in these stuctures. This an aircraft with a damage hardness of 6 psi would be damaged as an 18 psi target when in a hanger with hardness 3. The hardness factors for hangers and revetments is definable for each airbase by the user. Thus the user can define revetments at one base to be much stronger than at other bases, but leave the hanger factor at that airbase the default. Default values of 3 for hangered aircraft and 10 for revetments are used for airbases where the user hasn’t specified anything else.
There are several simplifying step taken in this algorithm to streamline airbase specifications. First, the size of the aircraft doesn’t matter. Second, survivability does not consider aircraft dispersal at the base (generally the lowest cost passive survivability measure). Third, there’s no priority consideration for the type of aircraft to be protected. If studies require these considerations, please let us know and we’ll update the algorithm.
Using this survivability measure, the actual aircraft destroyed is evaluated against the munitions load as follows:
The total number of targeted aircraft is estimated according to warhead guidance type. These estimates are as follows:
If the warhead is dropped in “blanket bombing” the total # aircraft affected is .02*number of bombs that fused.
If the warhead dropped is Unguided but dropped by a pilot, the total # of aircraft affected is .1 * number of bombs that fused.
If the warhead is guided, the total # of aircraft affected is .33 * number of bombs that fused.
If the warhead is smart, total # of aircraft affected is .67 * number of bombs that fused.
Finally, if the warhead is guided by external designation with deconfliction, the total # of aircraft affected is .8* number of bombs that fused.
When appropriate, the estimated total number of targeted aircraft is then adjusted according to submunitions as follows: The total number of bombs that fused is multiplied by (1-((submunitions*.2)*alog(urand_())). This adjustment is based upon the main unit targeting (hence the “1” factor) and a retargeting based upon a mean of one additional target per 5 submunitions taken from a negative exponential distribution.
Based upon this targeting factor, the expectation of targeting any specific aircraft is (targeting factor)/(# aircraft at the base). Note that this is a expectation, not a probability, so values of more than 1 are legal and represent retargeting.
Given these values, the algorithm loops through all aircraft at the base to determine their single strike probability of survivability as follows:
psone targeted bomb = ½ (/(kill radius/CEP)
where the kill radius is the specific munition kill radius (as defined in database prototyping) against the aircraft’s hardness (again, from database prototyping) times the survivability adjustment based on the likelihood the aircraft is in a hanger or revetment, as described above.
Given this, the PK for each aircraft is:
PK = (1.- psone targeted bomb) (Expected Targeting)
Each aircraft’s survivability is Monte Carlo’ed versus this calculation and kills are recorded.
Finally, the duration of airbase closure is computed as follows:
If more than two aircraft are killed:
Closure = (2*(# aircraft killed-2))*alog(urand_()) hours. This is a draw from a negative exponential distribution with a mean of two hours closure for each aircraft killed after the first one
Otherwise there’s no closure due to aircraft losses
In addition to this factor, an additional closure time is added it a munition designed for killing runways successfully hit the base. This closure is 12*(1+ alog(urand_()) hours This results in a closure of 12 hours plus an additional closure drawn from a negative exponential distribution with a mean of 12 hours. Thus the typical closure (I hesitate to say mean in a composite distribution) is 24 hours.
The computed airbase closure time is added to any preexisting closures to accommodate multiple strikes against the base.
Finally, it should be mentioned that the changeopstatus_ routine, in sim/objutil/asset_utilities.c, which is used to change asset operational status, has been changed to check that an event that re-opens an airbase does not attempt to re-open it before the damage has been cleaned up. This had to be done to recognize that an attack occurring while the airbase was being repaired would (usually) delay the reopening of the base and the original time to return operational would not be accurate.
A new “Special Characteristics” attribute has been added to the warhead definition. The only operational item for this is “anti runway”
The airbase specification GUI now incorporates information on the # of aircraft that can be parked in revetments and hangers at a base. To get to this interface edit the airbase in the Scenario Generation interface (shown on the left) and then select Special->Airbase Data”
No changes. The information for killed aircraft is as before. No data is currently stored on airbase closures.